This program is what I wrote for MATLAB Algorithmic Case in Rotman International Trading Competition 2016.
Due to the limited time, the code is a little scrappy.
Basically, this is a simple HFT strategy.
Main features are:
Take advantage of the order queuing and order book system of trading server by smart order management; Succeed in performing algorithm of market scenario recognition; Reduce risk position exposure by auto position balancing strategy; Accomplish fast open-position clearing algorithm.
This strategy is consistent on RIT simulation servers(Please refer to the png files for simulation performance).
FYI: RIT simulation servers executes a collection of orders(one queue) every 0.2s. During the time interval between two executions, all submitted orders are placed in a queue.